Fuqua Insights Podcast: What Happens to Innovation If Research Funding Gets Cut?

Tanner Morgan  0:03  
Welcome to Duke Fuqua insights, a Podcast, where we explore faculty research and the actionable takeaways for business leaders at every level. When universities get federal research grants for things like clinical trials for cancer treatments or understanding the health risks of hurricane flooding. A big chunk doesn't actually go to lab work. It goes to keeping the lights on, maintaining equipment and handling compliance paperwork. Those expenses are called indirect costs. Today, we're getting a behind the scenes look at how the US government funds scientific research. My name is Tanner Morgan, a recent Fuqua MBA graduate, and I'm joined by Fuqua Professor Daniel Gross. He's an expert on the economics of innovation, science policy and how institutions shape technological progress. In this episode, he helps us unpack how indirect cost funding works, why it matters, and what might happen if it were to suddenly disappear? Professor, thanks for joining me today.

Daniel Gross  1:04  
Tanner, thank you so much. It's a pleasure to be here and tell you about a topic that I've been thinking a lot about lately, an issue that's really lived in the shadows for decades as this arcane corner of federal science policy as it affects the broader innovation system in the US, but in the past few months has become incredibly relevant as different federal agencies are beginning to explore changing their policies around indirect cost recovery.

Tanner Morgan  1:29  
Yeah. So to start off, can you please walk us through what indirect cost recovery or ICR is and why it's necessary for scientific research funding in the United States.

Daniel Gross  1:41  
Great this is exactly the place we ought to start, and I think we should do so by first thinking about what the cost structure of research looks like. Scientific research generally incurs two categories of costs, fixed and variable. The variable costs are what we call the direct cost of research. These are project-specific and include things such as salaries in shares of full time equivalents or consumable supplies. Indirect costs, on the other hand, are costs that are not easily assigned to any one research project. Think about things like lab space, data and computing resources, the cost of complying with numerous regulatory requirements such as around safety and security, even heating and cooling costs, utilities costs, just like commercial firms, universities have fixed costs like these. This is the overhead cost of doing research. Now, indirect cost recovery, or ICR, as we'll call it, is historically how the US government has paid for the fixed costs of the research that it funds. For the last 60 years, the way this has been done is by paying universities and other research institutions an institution-specific fixed percentage increment over the direct cost of the research it supports, and this increment is known as an indirect cost rate. ICR rates are negotiated between the federal government and federally funded research institutions every few years, usually supported by many 100, hundreds of pages of documentation through this carefully audited and accounted process. But this rate is what is designed to really compensate federally funded institutions for their investments in infrastructure and in other overhead expenses that support the research they do, particularly that they do for the federal government. And without this indirect cost recovery, universities would have to pay for this fixed cost themselves, and many simply either couldn't afford to do so or wouldn't be attracted to, drawn to doing research for the federal government, the kinds of science that the public seeks these universities to undertake.

Tanner Morgan  3:41  
So you've written that the United States doesn't have a national university system like many other countries around the globe. How does that shape the way we fund research here in this country, particularly through indirect cost recovery? 

Daniel Gross  3:54  
Great, that's a good question. And when you think about what research and innovation systems look like around the world, many countries actually do have, as you said, national university systems where their budgets come directly from the government. That's not the case in the US. The US has historically had a much more decentralized system, kind of mimicking the more federalized structure of state and local government in the US too, not to mention a whole category of private institutions, many of which are among the most productive and successful in the world. In this context, the US government primarily funds research by providing grants for individual research projects, rather than block grants for institutions or for infrastructure specifically. And indirect cost recovery has become the key mechanism for helping institutions recover the full cost of research as it's paid over their cost of each individual grant. So ICR is by a large margin, the main way that infrastructure and other fixed costs that are necessary to conduct modern research are compensated. And it works like this -- just to put a little bit more meat on the bone of what it is we're talking about -- look, Duke University has a 61% indirect cost rate, and that means that if I propose a grant to the NSF for, let's say, $100,000 of direct cost that might pay for, for example, data might pay for OpenAI API credits, it might pay for research staff salaries and other things like that. I would budget in another $61,000 on top of that $100,000 for indirect cost recovery. Now it turns out Duke doesn't actually collect 61 cents over every dollar of direct costs, which is part of why ICR is very misunderstood. But we're gonna get into that, so I'll save the explanation of that for perhaps a few moments from now.

Tanner Morgan  5:47  
Yeah, so you reference that these negotiated ICR rates, and one of your big findings is that there's a huge gap between these negotiated rates and what the universities actually get paid. Why is there such a big difference, and what are the consequences of that?

Daniel Gross  6:03  
All right, there is so much to unpack here. I would almost not even know where to begin. But in order to get to this finding, I think we first have to think through and walk through what it is that we put together evidence-wise, and so I’m going to tell you what we're working with. And so if you go back… so really… let me get back to the beginning of this project. This project began in February of 2025, just a few months ago, when the NIH announced a sudden change in policy that was going to reduce its indirect cost recovery rate that it pays to all universities and the research it funds at universities to 15% -- and we'll get into that, I think, in more detail, in a little bit, what the ramifications of that might be. But when this guidance came out, a couple of co-authors and I had a phone call about other work that night, and we saw this and we thought, you know, this issue isn't really that well understood. Maybe we can say more about it, and maybe we can say more about it using data. And that’s what we set out to do. And so to better understand indirect cost funding, we gathered data on around 350 universities and other research institutions. Think about, for example, medical centers, medical schools, independent institutions like Scripps, or let's say, Dana Farber, and together, the set that we collected account for around 85% of total NIH research funding over the past 20 years. And for those institutions, we looked at both their negotiated ICR rates, but also what we call their “effective ICR rates”, which is the amount of indirect cost funding the institution actually receives over every dollar it spends on the direct cost of the research it performs for NIH. So what we need is data sources for both of these, both of these measures, the negotiated rates and the effective ICR rates. The negotiated rates we collect from directly from the institutional cost agreements, indirect cost agreements that these institutions enter into with the federal government. These are, again, as I said before, entered into every few years as a result of a complex negotiated process. But they're generally for most universities, public, and other institutions for many also public, and so we pulled data from these. And right alongside that, we go and we collect data from the NIH itself on its grant funding. And using NIH grant data, we calculate our so-called effective rates, defined as total NIH funding for indirect costs divided by total funding for direct costs for an institution in a given year. And our key finding when we compare these two rates was the following, we see that negotiated rates have been rising over a long period of time. Think back, let's say to 1980 we go back again further, but for our purposes, let's just take it back 40,45, years. When you think about what the rate consists of, I won't get into too much of the mechanics of how negotiated rates are computed, but suffice to say, in short, that these negotiated rates essentially reflect universities and other institutions recent indirect costs that are assignable to federally sponsored research divided by a cost base, a direct cost base associated with federally sponsored research too. And so you divide one way or the other, and you get a rate. And the fact that this rate has been going up over time has been triggering growing concerns that really the overhead costs of research have been growing and growing inexorably and potentially due to administrative bloat. I think that's kind of the the intuition that a lot of the public, or at least the policy making world, gets from this. On the other hand, however, if you actually look at the “effective” ICR funding, that is to say, what's being paid out, which you get from the NIH data itself, it turns out that effective ICR rates are much, much lower than negotiated rates, and moreover, they haven't really changed in 40 years. If you look at NIH as annual budget, the share of total grant funding that goes to indirect costs has been roughly constant for the last 45 years. And so even though many institutions have these negotiated rates between 50 and 70% over every direct cost dollar, what they're actually getting is much lower, more like 30 to 50% and so then look, we arrived at this puzzle, or this question, how do you explain this paradox of one rate that tells you one thing, another rate that tells you another thing. What we figured out, what the evidence generally points to, is a subtle but very crucial distinction between them. Negotiated rates are calculated off of a modified direct cost base. That modification reduces the direct cost base off of which those rates are both computed and applied. The idea being that there are certain direct costs that the government has decided we're not going to provide indirect cost recovery over. But that means the consequence of that is that for a given level of overhead at any particular institution, when the direct cost base off of which you can collect recovery or funding for that overhead, when that base declines, the rate at which you collect Indirect Cost Recovery has to go up in order to compensate in order to continue funding the same set of overhead costs. You end up with this mechanical relationship, the direct cost base that enters the denominator of this rate is going down, or things are holding it back, and the overhead itself that a university is incurring is not changing as a result, the negotiated rate is going up. The effective rates are computed off of not this modified cost base, but really the total direct cost funding that a university gets, and as a result, that stays unchanged. It comes down to really to accounting, and it wasn't until this paper that I came to really appreciate how much surprise we can find in accounting. 

Tanner Morgan  12:15  
Well, professor, you spoke about how indirect cost recovery is under a spotlight. It's in the news, as the NIH has recently proposed replacing negotiated rates with a flat 15% rate. For people outside of academia, why should the broader public care about something as technical as an ICR? What's at stake for them? 

Daniel Gross  12:38  
Great question. You know, I haven't lost you already right? After that incredibly ultimately detailed explanation of what ICR is and how it works, and why we're seeing the patterns we're seeing, and why this matters to begin with, the truth is, the devil of this stuff is in the details. At a higher level, ook, think about what ICR actually supports. A flat rate like the 15% the NIH has attempted to impose it's currently held up in litigation in federal courts and other federal funding agencies like the NSF, the National Science Foundation that is, the Department of Energy, the Department of Defense, they're looking to make the same changes. What this will do is reduce the funding that institutions get for the research they're performing for the public. And in this case, the particular case we're looking at is NIH specifically. And so what a 15% rate would do is reduce most institutions NIH funding by around 15 to 20% of what they were previously receiving on an annual basis. And that means that several institutions would experience declines in revenue of over $100 million per year, and in any case, for even other institutions, though, that all declines are smaller as a share of their total budgets, they're still fairly large as a result of this policy alone. So what? Right? Maybe these are taxpayer savings, but what this actually ends up doing as well is reducing incentives for research, especially incentives for high fixed cost but high potential research, for example, research that might require gene sequencing centers or imaging facilities or primate research facilities, where you have these model organisms that it's not testing in humans, but rather close substitutes, or bio containment labs or or clinical trial infrastructure, these things that that support high impact research but incur high fixed costs. And bigger picture, I think one has to also think what's the value of this research to the public in the US and also around the world. But I think broadly US policy is concerned with US taxpayers, US public benefits. And what this risks doing is upending a system that has for several decades been an engine of not only economic progress, but also health improvement in the US. And the the fruits of these investments are really what's at stake. And perhaps I can get into the details of those fruits in just a little bit too.

Tanner Morgan  15:10  
So your analysis has shown that some of the institutions hit hardest by this proposed change are the ones that are most connected to commercial innovation, things like new drugs, patents or startups. Can you walk us through that finding and what some of the implications of it are? 

Daniel Gross  15:27  
I mean, this is where the rubber really starts to meet the road, right? It's not just about, how are we supporting scientists and science and the work being done in the labs, but really, how does this come out of the lab and into the field and ultimately into the public. So here we collected additional data on NIH funded science, private sector patents that cite that science and new drugs that were developed over the last 20 years to explore how NIH supported research, scientific research, connects to US biomedical innovation. And what we found here is that institutions that are most likely to lose funding, and particularly those that lose the most funding under a 15% ICR rate, are also those whose research is most often cited by companies creating commercial products. For example, institutions that would face a 10% larger decline in NIH funding under this policy, on average, their NIH funded science is cited by 30% more private sector patents, and those patents are linked to 50% more commercial value. Similarly, of universities that had patents on at least two FDA approved drugs in the last 20 years, all of them had substantially higher ICR rates, both negotiated and effective, than the proposed 15% cap, and so these universities would lose significant funding under the 15% ICR rate proposed. And what's at stake then, potentially, are the fruits of that innovative labor

Tanner Morgan  16:59  
Your research has found that other countries fund research very differently. Could a shift in ICR policy affect our global standing in science and innovation? 

Daniel Gross  17:08  
I think that's one of the concerns. And look, we don't have a perfect counterfactual. We don't know exactly how the system will adapt to a change like this until it takes place. There's been nothing quite this systemic, especially around ICR policy in quite some time, indirect cost recovery based on institution specific negotiated rates has been in place for 60 years, since the 1960s and indirect cost recovery generally has been a feature of federal research funding since the 1940s. So what this would do remains to be seen. That said, certainly there's a lot at stake. There are big risks. Even if funding freed up by the cap were reinvested in two additional research grants, particularly into direct costs for other other research, it would still distort the types of research pursued. It would steer solutions away from high, fixed cost projects, like I said before, and on the other hand, if the funding were simply cut rather than reinvested, the damage to US innovation, especially in biomedicine, which is heavily science dependent, could be significant. It's this science that underpins American competitive advantage in a whole range of science-based technological industries and the pharmaceutical industry and other biomedical industries like devices being high on that list.

Tanner Morgan  18:35  
So you've mentioned that ICR often draws criticism people think it's bloated or lacks transparency. How would you respond to that? And what reforms do you think might strike a better balance? 

Daniel Gross  18:49  
I think we have to admit that some of these critiques are well founded. Indirect Cost Recovery has been a subject of policy debate for decades. There are flaws in the system, and I'll come back to think about what some alternatives perhaps could look like too, and what trade offs they present. But let's just first embrace that the current system isn't perfect. It's complicated. The process of simply producing a proposal for a new negotiated rate every two to four years that universities and other institutions have to undertake, that can cost millions of dollars. It's very complicated, and not to mention the whole process is opaque and hard for most lay people to understand. Took us a long time to even wrap our heads around it in order to try to communicate it in our own research. At the same time, I think it's also important to recognize that ICR doesn't just fund administrative costs. For that matter, it mostly funds facilities. The share of ICR that goes to administrative costs is actually legally capped, and has been since 1991 and most universities have been at that cap for many years. Basically the ICR rate that can be produced or attributed to administrative costs is limited at 26% and that hasn't changed. Most universities are at that cap year in and year out. And so the idea that ICR is funding growing administrative bloat, for that matter, has to be factually incorrect. And moreover, many of these costs are incurred in service of complying with a growing number of rules, federal rules that regulate federally sponsored research, including the cost of administering ICR to begin with. And so, because of that, and because of this cap that I just mentioned, actually, even now, universities appear to under recover a meaningful portion of the true and direct costs associated with federally sponsored research. We have evidence of that, examples of that from a number of universities. In any case, there is almost certainly room for improvement. The current system has its own flaws. Some of those flaws have to do with the incentive system that it creates when ICR, when indirect costs and infrastructure investments can be recovered through renegotiated ICR rates every few years -- that can encourage growth of the enterprise investments in new buildings, new facilities that may be marginally less productive, or that otherwise an institution might not make and so one can imagine that there are other incentive structures that might rein in some of those choices, and one of them is a simple, flat 15% rate. So think about what, what a 15% rate does. It's certainly simplifying. It forces more cost sharing, and so, in turn, maybe some more judicious investment choices on the part of universities and other institutions. But it's so low that it also risks, and not only that, but in my view, is likely to make a lot of high value life improving and saving research just economically not viable. But in between, there are other options. I've been thinking about this, and I've been asked this over time too. If you had to put my feet to the fire, make me tell you what I think a good alternative would be, what would I do if I were in the driver's seat? First of all, I would say I don't want to be in the driver's seat, because it's such a complex thing to resolve that. In some ways, who am I to make that choice? I think that choice is up to others. But if pressed to have a point of view, I can see two ways forward. One way would be a system where we allow rates to vary for different types of institutions. Maybe there's one for universities, one for medical schools, one for independent research institutions and so on. And we could set these rates around their historical average effective rates, which, as we've shown in our work, haven't really changed in many decades. And we would ditch this complicated, modified direct cost accounting that changes what goes into the denominator, which then influences the negotiated rates we see today. Forget that. Keep it as simple as possible. Pay ICR over the full set of direct costs of research, which is what the effective rates are essentially defined over. And so that's one way. The other approach I could see is to try to shift more of these costs, as we call it above the line, into direct cost accounting, to try to apportion some of these fixed costs to specific research projects that can't be done with all fixed costs, but but perhaps can be done better with some and so the system can be moved more in this direction. And so incidentally, or perhaps not incidentally, these are similar to the ideas that a recent working group on the issue, that consists of administrators from several leading US universities, called the Joint Associations Group or JAG. These are similar to the proposals that this JAG group put out to the public about a week ago, and as far as I understand partly based on our own work.

Tanner Morgan  23:53  
So looking ahead, how should students or future business leaders think about the role of infrastructure, incentives and policy in driving innovation? 

Daniel Gross  24:03  
It's a great question, and this can actually turn into a big picture question. So I'm going to take this up one level to think about the role of federal funding in the US innovation system, not only for infrastructure, but for other activities too. To me, it's useful to embrace, or at least recognize that the federal government, universities and industry have been working together for generations to advance science and innovation in pursuit of really three ends. One is bolstering national security. Another is supporting and promoting economic prosperity and growth, and a third is improving health. Of course, the NIH is the engine behind US biomedicine, the largest single funder of biomedical research in the US and the world, for that matter, too, but this partnership has brought us everything from antibiotics to AI. It's won wars, it's ended pandemics. Think about where we are today compared to where we were a century ago. A century ago, the US government had very little role in supporting science, and science was much smaller in scale, and in its impact than it is today. This changed in World War Two, when exigency really the urgency of the moment inspired a collaboration between the government, firms and universities to develop science and technology for war to solve problems that just needed urgently solving. And so it was in this moment of need that a new relationship, this new collaborative partnership was formed, and its success in that moment inspired its not only continuation, but its expansion over the next several decades. And so here's how it looks today. The US innovation system now has a pretty sharp division of labor between universities and firms. Universities do the science and firms find way to put it to work, to turn that into useful technologies, whether it's drugs, whether it's electronic devices or anything else, and this innovation makes our lives safer, healthier, longer, more prosperous. And in the background of this flywheel sits the federal government, which funds most of the science. And that's because if it didn't, history and economics both point us to believe, give us reasons to think that a lot of the science really wouldn't get done, especially risky science, science which doesn't have immediate application, science with large upfront costs, science which you can't take out intellectual property on and that's a lot of the modern scientific enterprise. And so think about what this has given us. US life expectancy has increased more than 30 years over the past century. That improvement in life expectancy and health is estimated by some to be just as valuable as all economic growth over that same period. We now make in real terms in America, about twice as much at the median as we did in 1950, we're richer, we're more comfortable. A lot of this progress is attributable to science driven innovation, and so that all leads me to think that without science, our lives would be shorter and poorer. Without public funding, we wouldn't have much of that science. And right now then, with that in mind, that context, right now, we're at a moment, a juncture, really, where the partnership that has been allowing this flywheel to keep spinning and spinning faster that's been beneficial for so many decades is under new strain. It's become a bit more adversarial, particularly in government and universities, and that seems, from where I sit, to reflect a little bit of a shift in how the government views this partnership, something that was long treated as a public investment that's designed to benefit that ultimately benefits the American public, is now being treated a bit more instrumentally as a gift that supports university goals as opposed to public goals. And so we end up with this tension and this dance we see taking place in policy, in the courts, where we don't quite know how the game is going to end. But bigger picture, permanently cutting billions of dollars in research funding is probably going to be damaging. I think you know most, most folks can believe or at least come to agreement that there's a high likelihood that cutting this much funding from the US scientific enterprise can be very damaging to American innovation and ultimately, technological leadership. Because if you take out a three-legged stool like the one we're talking about, where you have government, industry and universities, and you knock out one of the legs, that stool is going to fall over. And so look, if you bring this back up, there's a reasonable case to be made that this partnership can use some rejuvenation, some reform. There are opportunities to make it more efficient, to make it better. But for a system that's been as productive as the US innovation system has been over the past, not just 20 years, but really 80 years, I would think that the right tool for the job is probably closer to a scalpel than it is to a sledgehammer. And with that in mind, I would encourage those who are listening to come away from this thinking about what kinds of adjustments to US federal research funding and research policy would be consistent with improving the system while maintaining and sustaining its successes.

Tanner Morgan  29:05  
Thanks for your time today, Professor.

Daniel Gross Thanks very much, Tanner.

(music)